CN111480276A

CN111480276A - Controlled energy storage balancing techniques

Info

Publication number: CN111480276A
Application number: CN201880069617.5A
Authority: CN
Inventors: 徐文泰; 常文天; 李传宏; 李振辉; 郭正朝; 罗文宗
Original assignee: LT Lighting Taiwan Corp
Current assignee: Lt Usa
Priority date: 2017-10-27
Filing date: 2018-05-31
Publication date: 2020-07-31
Also published as: CL2020001035A1; PH12020550460A1; JP2021501559A; CO2020005191A2; MX2020004130A; KR20200080239A; TW201918011A; WO2019083568A1; CA3077405A1; EP3701613A4; TWI677180B; EP3701613A1; BR112020007686A2; AU2018355031B2; JP7460527B2; KR102394942B1; AU2018355031A1; SA520411836B1; EA202091049A1

Abstract

An energy storage system includes an energy reservoir and a system controller. The energy reservoir is charged by DC energy from the DC energy source while releasing the DC energy to the DC/AC converter. The system controller regulates the amount of DC energy released from the energy reservoir to the DC/AC converter to bring the amount of DC energy charged to the energy reservoir near equilibrium. Since the charge and discharge are close to equilibrium, the size of the energy reservoir can be made very small relative to the amount of charge and discharge. This is advantageous in situations where the charge and discharge flow is high, as may be the case if the energy storage receives charge from all or most of the power plant, e.g. a solar power plant. With such a controller, it becomes technically feasible to use the energy reservoir even in the case where such a large current flows.

Description

Controlled energy storage balancing techniques

Background

Photovoltaic (PV) power plants generate electricity by converting solar energy into electricity. The generated electricity is then provided to the power grid. Solar energy sources (i.e., received solar rays) are characterized by having a time-varying intensity. Therefore, the PV generator in such PV power plants comprises a power generation optimization device (also referred to as "optimizer"). One type of optimizer is known as a "maximum power point tracker" (MPPT) (or "MPPT device") that tracks the instantaneous maximum power generation point (MPPP) voltage used by the MPPT device to control the operation of the PV power plant. This practice is referred to herein as the "blind MPPT configuration". MPPT devices are typically software or firmware; and tracks the time-varying voltage, resulting in maximum power generation from the time-varying solar energy source.

The subject matter claimed herein is not limited to implementations that solve any disadvantages or to implementations that operate only in environments such as those described above. Rather, this background is only provided to illustrate one exemplary technology area in which some embodiments described herein may be practiced.

Disclosure of Invention

Embodiments described herein are directed to an energy storage system including an energy reservoir and a system controller. The energy reservoir is charged by DC energy from the DC energy source while releasing the DC energy to the DC/AC converter. The system controller regulates the amount of DC energy released from the energy reservoir to the DC/AC converter to bring the amount of DC energy charged to the energy reservoir near equilibrium. Since the charge and discharge are close to equilibrium, the size of the energy reservoir can be made very small relative to the amount of charge and discharge. This is advantageous in situations where the charge and discharge flow is high, as may be the case if the energy storage receives charge from all or most of the power plant, e.g. a solar power plant. With such a controller, it becomes technically feasible to use the energy reservoir even in the case where such a large current flows.

The system controller includes a detection component, a determination component, and a delivery component. The detection component is configured to measure a level of stored energy in the energy reservoir. The determining means is configured to use the measured energy storage level of the energy level to evaluate whether an adjustment is to be made. The delivery component is configured to encode an instruction to perform the adjustment into an encoded message when the determination component determines that the adjustment is to be made, and is further configured to deliver the encoded message to the DC/AC converter.

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

Drawings

In order to describe the manner in which the above-recited and other advantages and features can be obtained, a more particular description of various embodiments will be rendered by reference to the appended drawings. Understanding that these drawings depict only sample embodiments and are not therefore to be considered to be limiting of its scope, embodiments will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:

figures 1A to 1C show block diagrams of various power plants in which decoupling means are used in conjunction with an energy store;

FIG. 2A shows a block diagram of a power plant arranged in an experiment and in which there are two AC power generating units arranged conventionally and having power and energy meters measuring the output of each power generating unit;

FIG. 2B shows a block diagram of the power plant of FIG. 2A after modification to include decoupling means and an energy reservoir and for verification of improved energy output to the power grid;

FIG. 3 shows a block diagram of a power plant in which there are two channels of power delivery, one channel invoking use of the energy reservoir and one not;

FIG. 4 shows a block diagram representing a power plant of the broader embodiment of FIG. 3;

FIG. 5 shows a block diagram of a power plant in which power is delivered via use of an energy store;

FIG. 6 shows a block diagram representing a power plant of the broader embodiment of FIG. 5;

FIG. 7 shows a block diagram of a power plant;

FIG. 8 illustrates a block diagram of a Maximum Energy Utilization Point Tracking (MEUPT) controller according to principles described herein; and

fig. 9 shows a block diagram of the MEUPT controller of fig. 8 in the case of a power plant.

Detailed Description

Patent publications US2016/0036232 and US2017/0149250a1 (the contents of which are incorporated herein by reference) disclose that practicing PV energy systems in a blind MPPT configuration achieves a suboptimal amount of power provided to the grid. These patent publications teach that, in order to efficiently extract electricity to utilize energy, one should match the characteristics of the energy extraction device to efficiently and effectively extract the generated electrical energy. Furthermore, these patent applications teach that the relevant devices should also be matched to condition and/or deliver the extracted electricity to efficiently utilize the energy.

These patent publications also emphasize the following facts: in addition to power generation, energy utilization efficiency is also indiscriminately dependent on power demand. Further, these patent publications teach that in any energy system, the typical power consumption is not necessarily equal to the power generation, even when the laws of conservation of energy and charge are adhered to.

Instead of using MPPT devices as optimizers for solar power plants, the cited patent publication suggests using "maximum energy utilization point trackers" or "MEUPT devices" as PV power plant optimizers. Such an optimizer will be referred to herein as a "MEUPT optimizer". According to the cited patent publication, the MEUPT optimizer is designed to capture its so-called "residual energy," which is defined as electrical energy that is produced but not extracted and/or delivered to the grid for utilization. This definition of "residual energy" is also used herein.

The MEUPT optimizer is further designed to temporarily store the captured remaining energy within the energy store; and then prepare and deliver the electrical energy to the grid for utilization. Thus, when the MEUPT optimizer is included, the electricity sales revenue of the PV power plant can be increased.

A first part: MEUPT optimizer functionality

According to the principles described in US2016/0036232 and US2017/0149250a1 ("cited patent publications"), the MEUPT optimizer of one embodiment disclosed herein includes a residual energy extractor, an energy reservoir, and a MEUPT controller. The MEUPT controller works in conjunction with the energy extractor and the DC/AC converter. The terms "power" and "energy" (although not exactly the same) are used interchangeably in the art. Thus, unless otherwise indicated, each term has the same meaning.

An energy extractor extracts an initial oscillating power train from the generated DC power. The extracted initial power chain conforms to AC grid requirements of the grid. In other words, the extracted initial power train has a time-varying sinusoidal voltage whose peak voltage conforms to the grid voltage range. Furthermore, the power (proportional to the square of the voltage) is taken as (sin)²(ω t) or cos²(ω t)) that is synchronized to the grid (has the same phase and the same frequency).

On the other hand, the residual energy extractor extracts the remaining oscillating power system obtained by subtracting the initial oscillating power system from the generated DC power. In other words, the remaining oscillating power train is the remaining oscillating power train remaining after providing the initial oscillating power train to the grid. The remaining oscillating power train has a 90 ° phase shift compared to the initial oscillating power train provided to the grid. Due to the 90 ° phase shift, the remaining oscillating power cannot be immediately converted to AC power for supply to the same grid. Therefore, an energy reservoir is used to temporarily store the remaining energy of the remaining oscillating power train. Thereafter, supplying the stored energy to the DC/AC converter; so that the stored residual energy can be converted into AC power that is synchronized (has the same phase and frequency) with the same grid.

The MEUPT controller measures the energy level of the energy storage; estimating an amount of energy that can be extracted in the energy reservoir; and deliver this information to the associated DC/AC converter so that the energy can be extracted by the DC/AC converter. The DC/AC converter then extracts the stored energy from the energy reservoir, converts it to AC power in the form of a suitable pulsed power train, and provides the AC power to the grid. Thus, a PV power plant, when including a MEUPT optimizer, can provide almost all of the generated electrical energy to the grid. In contrast, a PV power plant according to the cited patent publication can only provide less than half of the generated power/energy to the grid without a MEUPT optimizer.

A second part: retrofitting conventional PV Power stations with MEUPT

The power rating of a solar power plant is typically in units of a certain number of Megawatts (MW). Conventionally, when the rated power of a solar power plant is stated to be x MW (where x is some positive number), this means that the sum of the DC power production rated powers of all the solar strings is x MW. Such a conventional solar power plant also has a three-phase DC/AC converter, the manufacturer of which states that the total DC/AC conversion capacity is not greater thanx MW. This principle summarizes the operation of conventional power plants in accordance with conventional MPPT practice.

Word changing deviceA conventional PV power station with a power rating of x MW comprisesx MWStrings of PV solar panels that convert solar energy to DC electricity. The generated DC power is then extracted by a three-phase DC/AC converter and converted to suitable AC power that meets all AC power requirements of the grid and is then provided to the grid. This AC power provided to the grid is also referred to herein as the "primary oscillating power train". Recall that the manufacturer claims that the total DC/AC conversion capability of the DC/AC converter is no greater thanx MWThis is the total amount of DC generation capacity of the installed solar panels stated by the solar panel industry.

According to the description of the cited patent publications US2016/0036232 and US2017/0149250a1, there is a remaining oscillating power train that is generated when the initial oscillating power train (extracted by the energy extractor) is subtracted from the total DC power generated by the solar panel string. In other words, the power train is the remaining oscillating power train, which has about a 90 ° phase difference from the initial oscillating power train extracted by the energy extractor and provided to the grid.

Since the remaining oscillating power train is about 90 ° out of phase with the grid, it cannot be directly regulated and converted to AC power and provided to the same grid. According to the principles disclosed in the cited patent publication, the energy reservoir temporarily stores energy (representing the remaining energy when stored) containing the remaining oscillating power train that is 90 ° out of phase. After storing the residual energy in the energy reservoir, the residual energy may be used as DC energy that may be supplied to the DC/AC converter. This remaining energy may then be converted to AC power that meets all grid requirements (including synchronization with the grid), such that the resulting AC power may be provided to the same grid.

And a third part: preventing energy leakage from an energy reservoir

Before elaborating on energy reservoir design considerations for the MEUPT optimizer, an important issue is addressed herein first. In particular, a string of solar panels may have a very high resistance at dusk, but a string of solar panels may conduct a large amount of current in either direction at noon when the sun is intense. Therefore, the electric energy stored in the energy reservoir may leak and heat the solar cell panel during the daytime. Thus, a decoupling diode may be added to each of the strings of solar panels such that electrical energy may flow from each string of solar panels to charge the energy reservoir, but energy in the energy reservoir cannot flow back from the energy reservoir into the string of solar panels. Various energy reservoir systems that achieve this decoupling will now be described with reference to fig. 1A, 1B, and 1C.

The fourth part: design considerations for energy storage

Fig. 1A depicts a block diagram showing an energy reservoir 1300A designed to temporarily store the remaining power resulting from subtracting the power drawn by the DC/AC converter 1200A from the power flow generated by the set of solar strings 1100A when the DC/AC converter 1200A converts the power to AC power. AC power is provided to AC power grid 1600A through transformer 1500A. The energy reservoir 1300A receives the remaining oscillating power train through decoupling diode bank 1400A. In one example, the energy reservoir 1300A is designed to temporarily store the remaining energy of a 1MW PV power plant for 2 minutes.

By way of example only, assume that the primary energy source can maintain a constant intensity (and that the power production of the PV string 1100A is maintained to allow for a constant 1MW generator power production) for 2 minutes. For the following analysis, both the initial oscillating power train and the remaining oscillating power train have the same repetitive pattern, but have a phase difference of 90 degrees. First, consider how a brute force design energy reservoir can be used. It should be remembered that the purpose of the energy reservoir is to temporarily store the remaining energy so that the DC/AC converter can convert this stored energy later.

As discussed in the referenced patent publication, for a typical conventional PV power plant, the estimated ratio of residual energy to generated DC power exceeds 0.5. For analysis, assume a PV power plant has a string of 1MW PV solar panels; and converting the DC power to AC power for provision to a grid at 50 hz and with 380V three phase AC power present_ACLine voltage. In thatIn this case, the duration of one power cycle is equal to about 0.01 seconds and the total phase current is up to 1,000,000/(380/1.732), where 1.732 is a square root value of 3. This ratio is the ratio of the peak voltage to the line voltage (the line phase voltage or "phase voltage" in three-phase AC power). Storing the charge associated with the remaining energy in the power cycle of the power plant would require an equivalent charge capacity of about 8V faradays (0.5 x 0.01 x 1,000,000/(380/1.732)), where "V" is the voltage difference of the designed energy reservoir before and after charging.

To maximize the energy utilization of the PV plant, in some embodiments, the operating voltage of the MEUPT optimizer should be within 75% of the PV maximum power generation voltage. In other words, a voltage range of 75% of maximum power production will be observed in those embodiments of the MEUPT optimizer. The measured I-V data indicates that the range is typically about 80 volts. When this voltage range is selected as the charge/discharge voltage range for the energy reservoir (i.e., V-80 volts), the charge capacity of the energy reservoir is about 0.1 faradays per MW per power cycle (where a power cycle lasts 0.01 seconds).

If the design consideration is to store the maximum amount of residual energy accumulated over two (2) minutes, the required equivalent charge capacity is equal to 1200 farads (100 x 120 x 0.1) for a 1MW PV power plant. This required equivalent charge capacity is referred to herein as the "full maximum charge capacity" and the amount of energy stored by the energy reservoir associated therewith is referred to as the "full maximum energy reservoir capacity" or "full maximum remaining energy".

If only thin film capacitors are used to meet this required charge capacity, the volume of a set of thin film capacitors required to achieve this charge capacity would be very large and the capital cost would be very expensive. Therefore, it is impractical to design such a reservoir that includes only film capacitors.

As a break in this brute force design, a faraday (e.g., a battery) can be incorporated into the design to reduce volume and size. Careful analysis by the inventors has shown that the required charge capacitance is indeed technically manageable for a power reservoir having a thin film capacitor and a faraday arrangement. However, the cost of such an energy storage is still too high to be beneficial unless the price of the battery can be reduced by at least three times while maintaining the same performance.

The use of electrolytic capacitors can greatly reduce the required capital cost. However, this will increase the operating costs, since the lifetime of such capacitors is relatively short. Therefore, at present, the use of electrolytic capacitors is not feasible. Thus, a brute force approach does not enable an economically advantageous design with the required full rated maximum energy reservoir capacity.

The principles described herein address this problem using the following facts observed by the inventors:

(1) most existing DC/AC converters can easily step up or down power by 3% in one second; and also the existing 500kW DC/AC converters can easily rise or fall by more than 10kW in one second during operation.

(2) As a rough observation, a typical 1MW PV power plant starts power production from zero power every morning and rarely ramps up its power production at a rate in excess of 10 kW/sec in its normal daily operation.

(3) PV power stations at MW level (rated power greater than 1MW) may occasionally experience a rise rate of greater than 10kW per second during a brief power burst. However, the energy contained in this short burst (or even bursts above 100kW per second) is negligible when compared to the total daily energy produced in a MW level power station.

From these three facts, the inventors determined that (1) the power generation of each of the strings of solar panels starts from zero every morning; and (2) PV generators cannot immediately generate full power. Therefore, the remaining oscillating power does not immediately rise to its maximum value. In other words, the remaining oscillating power train is typically increased much more than the rise rate of the DC/AC converter. Furthermore, for PV stations rated at 1MW or higher, the amount of energy in any short rise burst is not a significant issue in energy harvesting.

Thus, instead of designing an energy reservoir capable of storing the full maximum amount of remaining energy, the principles described herein suggest designing an energy reservoir for storing a net energy equal to the difference between (say 2 minutes) the remaining energy input into the energy reservoir and the energy extracted from the energy reservoir by the DC/AC converter. This amount of energy is referred to herein as the "maximum differential residual energy". The amount of this maximum differential residual energy is much less than the full maximum residual energy. Therefore, it is easier to design such a smaller energy reservoir; this is technically manageable and cost-effective.

Fig. 1B depicts a block diagram symbolically illustrating an energy reservoir 1300B, the energy reservoir 1300B storing the surplus power resulting from the power flow generated by a set of solar strings 1100B minus the power drawn by the DC/AC converter. At the same time, another DC/AC converter 1202B is directed by MEUPT controller 1310B to receive approximately the same amount of DC energy (including excess power) from energy reservoir 1300B. Both DC/

AC converters

1201B and 1202B simultaneously convert the received DC energy to AC power and provide the AC power to the same grid 1600B through the same transformer 1500B. In so doing, the net energy storage burden of the energy reservoir 1300B may be reduced to a very small capacity when compared to the remaining power of the energy reservoir 1300A depicted in fig. 1A.

FIG. 1C depicts a configuration modified from that depicted in FIG. 1B but having approximately the same performance as the configuration depicted in FIG. 1B. As depicted in fig. 1C, energy reservoir 1300C stores the DC power flow generated by PV solar string 1100C through diode bank 1400C. The two DC/

AC converters

1201C and 1202C are directed by the MEUPT controller 1310C to receive (in general) about the same total DC power from the energy reservoir 1300C, the amount of total DC power being about equal to the DC energy input produced by the PV string. Thus, there is only a very small net balance of power input in the input and output of the energy reservoir 1300C. 1201C and 1202C both simultaneously convert the received DC power to AC power that is provided to the same grid 1600C through the same transformer 1500C.

In summary, as depicted in fig. 1B (when properly decoupled), the energy reservoir can extract and store the remaining energy in the form of the remaining oscillating power train, which is obtained after extraction of the generated DC power by the energy extractor (which can be built-in as a module of the DC/AC converter 1201B). The other DC/AC converter 1202B is designed to draw approximately the same amount of energy from the energy reservoir 1300B to reduce the net amount of remaining energy stored into the energy reservoir. Therefore, a relatively small energy reservoir is sufficient.

Also as depicted in fig. 1C (when properly decoupled), the energy reservoir 1300C may receive all generated DC power from the PV string 1100C. Then, the oscillating power trains are extracted by the DC/

AC converters

1201C and 1202C, while the surplus energy (surplus power) is also implicitly stored in the energy accumulator 1300C in the form of the remaining oscillating power trains that are 90 ° out of phase. It can be seen that this remaining energy is also implicitly automatically extracted and stored into the energy reservoir 1300C.

Applying either of the designs depicted in fig. IB (or fig. 1C), the designed energy reservoir can be used as an energy reservoir for the MEUPT optimizer, which temporarily stores a small amount of net residual energy that is 90 ° out of phase. The arduous task of energy reservoir design is now shifted to the task of designing an appropriate MEUPT controller.

The fifth part is that: essential functions of MEUPT controller

The controller should be able to direct the associated DC/AC converter to always draw an appropriate amount of energy from the energy reservoir, which energy is substantially equal to the amount of remaining power charged in the energy reservoir. This minimizes the net energy stored into the energy reservoir; and maintain sufficient balanced energy storage in the energy reservoir to stabilize system operation. In doing so, the energy reservoir need only store (or provide) the energy difference between the charging surplus power and the power drawn by the DC/AC converter for a short duration.

The energy difference can be designed to be small by a capable controller. The duration can be designed to be long enough to allow the DC/AC converter to ramp up or down when the remaining energy is matched; and short enough to significantly reduce the capacity of the energy reservoir while still maintaining system operation stability. Thus, the estimated energy reservoir capacity may be reduced to less than 0.001 times the full maximum remaining energy. The capacity per 1MV PV power plant is less than 2 faradays; even with the use of thin film capacitors, charge capacity can be managed. Examples of suitable MEUPT controllers are described below with respect to sections twelfth through fourteenth.

A sixth part: capacitor/battery combined energy storage device

Another problem is that good film capacitors can last 10 to 15 years while still retaining more than 80% of their original capacitance; while a good battery may last less than 5 years and after that its charge capacity is about 70%. Therefore, careful design balancing is recommended to optimize economic costs. Furthermore, the amount of energy in the energy reservoir should be large enough to keep the operation stable at all times. Design simulations show that with current prices of film capacitors and batteries, a typical 20 year optimal energy reservoir design is that of an automated battery string of about 50 amp-hours with a suitable operating voltage using 0.1 to 1 faradaic film capacitor combinations for a 1MW PV station.

A seventh part: preventing mutual power annihilation in PV strings

As described above, the decoupling techniques applied in fig. 1B and 1C allow the solar panel string to charge the energy reservoir; but prevents power from flowing from the energy reservoir back to the PV solar cell string. This technique not only prevents energy leakage from the energy reservoir through the string of PV solar panels, but also prevents the phenomena found by the inventors when the set of decoupling diodes is suitably applied. This phenomenon is referred to herein as "mutual power annihilation phenomenon", or "power annihilation phenomenon" in the PV string.

This occurs when several PV strings connected in parallel collect the generated power. This phenomenon is particularly pronounced when PV strings connected in parallel have very different I-V characteristics, photoelectric conversion efficiency and/or maximum power generation voltage.

For example, when less than all of the solar panels in less than all of the strings are shadowed, the strings within the shadow will have lower photoelectric conversion efficiency than the strings outside the shadow. In other words, due to the difference in shadow casting, the solar strings will have very different I-V characteristics even at the same time of day. When these solar strings are connected in parallel, the high efficiency string may release a portion of the power it generates to the lower efficiency solar string, thereby interrupting the power generation of the PV solar string. The inventors have experimentally confirmed this phenomenon. Experiments have also shown that this phenomenon can be prevented when the PV solar string is properly decoupled.

Furthermore, when PV strings connected in parallel have very different maximum power generation voltages, power annihilation phenomena also occur. For example, assume that there are two strings of solar panels connected in parallel, one with 15 string solar panels and the other with 19 string solar panels. The power generated in the string of 19 panels will definitely flow through the string of 15 panels and a power annihilation phenomenon will occur. Experiments have shown that the power received from the two strings connected in parallel above can be reduced to less than half the power generated by a string with 19 panels alone. When properly decoupled, the power received from the upper two parallel connected strings can be restored to 1.53 times the power generated by the string with 19 panels alone. The experiments described above show that: (a) mutual electric annihilation phenomenon does exist; and (b) proper decoupling techniques can prevent this phenomenon.

In another experiment, a PV power plant was arranged with two power generation units; each unit comprised 85 solar panels of the same manufacturer and model. Each of the two power generation units is configured with five (5) PV strings connected in parallel to collect the generated DC energy. Two PV strings are configured with 15 series-connected panels, two strings are configured with 17 series-connected panels, and the other string is configured with 21 series-connected panels. The maximum power generation voltage in these 10 strings ranges from the lowest 420 volts to the highest 610 volts, as measured at noon on a clear sky, respectively. Thus, under the same clear sky, these parallel-connected PV solar strings have very different maximum power generation voltages.

Each of the power generation units converts the collected DC power to AC power via a different DC/AC converter. To measure the energy and power generated in each generation unit, kilowatt-hour meters and wattmeters are connected to the AC output of each of the DC/AC converters in each generation unit. These units are then connected to a transformer to provide AC power to the grid. By having 72 identical readings from both watt-meters over a 36 day period, and by having the same readings from both kilowatt-hour meters at the end of the 36 day period, it can be confirmed that all of the components in both power generation units (including both sets of meters) are substantially identical.

Then, one power generation unit was modified to be configured with 4 strings of 21 panels (and 1 panel not in use); while the other power generating units remain unchanged from the 5 strings described above. At noon and during sunny days, the measured power production of the modified power generation unit is typically greater than 4.1 times the power production of the other power generation units. The sixty (60) days of cumulative energy provided is then measured, which is derived from the readings of the two kilowatt-hour meters. The modified power generation unit provides 3.38 times more energy to the grid than the unmodified power generation unit. The above experiments clearly and unambiguously demonstrate that there is indeed a mutual power annihilation phenomenon in parallel-connected strings of PV; especially for strings with very different I-V characteristics or very different maximum power voltages.

In summary, appropriate decoupling techniques in accordance with the principles described herein can prevent energy from leaking from the energy reservoir through the solar cell string; and also to prevent the phenomenon of mutual power annihilation in the PV strings found.

The eighth part: experiments to demonstrate the existence of residual energy

Before describing the design of the MEUPT optimizer, this section describes experiments to unambiguously demonstrate the presence of residual energy in such PV power plants; this is predicted by the cited patent publications US2016/0036232 and US2017/0149250a 1. Restated, the cited patent publication defines residual energy as electrical energy that is generated but not extracted and/or utilized prior to conversion to heat. Specifically, in a PV power plant, "residual energy" includes residual electrical energy present after the generated DC energy is extracted and converted into AC power by a three-phase DC/AC converter. The MEUPT optimizer may be designed to capture/utilize this remaining electrical energy, i.e., the remaining energy. The experimental setup and the stepwise execution of the experiment are described below.

Fig. 2A depicts a starting setup of a PV power plant 2000A comprising 2 AC

power generating units

2100A and 2200A. Each of AC

power generation units

2100A and 2200A practices a blind MPPT configuration; and provides three-phase AC power to grid 2600A. The AC power generation unit 2100A includes a DC generator 2110A and a three-phase DC/AC (15kW) converter 2130A. The AC power generation unit 2200A includes a DC generator 2220A and a three-phase DC/AC (15kW) converter 2230A. The generator 2110A generates DC power using 2

PV strings

2111A and 2112A connected in parallel. The generator 2220A uses the other 2 parallel-connected

solar strings

2221A and 2222A to generate DC power. Each of the 4 PV strings includes 25 solar panels connected in series; each panel is capable of generating 250W of power at noon and during clear sky.

DC generator 2110A supplies DC power to three-phase DC/AC converter 2130A; and the DC generator 2220A supplies DC power to the three-phase DC/AC converter 2230A. Then, the two

converters

2130A and 2230A convert the supplied DC power into three-phase AC power. In an experiment, the AC output power of the

power generation units

2100A and 2200A was measured by two three-phase AC wattmeters (in kW) 2351A and 2352A, respectively. The AC energy production (in kW x hours) of the two

power generation units

2100A and 2200A was also measured by two kilowatt-hour meters 2361A and 2362A, respectively. The generated three-phase AC power is then provided to grid 2600A via transformer 2500A. The PV power plant has been operated; and energy generation of two AC

power generation units

2100A and 2200A for 7 days was measured.

During this time period, the readings of the two kilowatt-hour meters per day show equal values. This provides a high degree of confidence that all elements of the two

power generation units

2100A and 2200A (including both sets of measurement instruments) are substantially identical. After this step, one of the two AC power generating units 2200A remains unchanged, while the other AC power generating unit 2100A is modified to have a different configuration 2100B as depicted on the left hand side of fig. 2B.

The power generation unit 2200B of fig. 2B is an unmodified power generation unit 2200A of fig. 2A. Also,

elements

2351B, 2361B, 2352B, 2362B, 2500B, 2600B of fig. 2B are the same as

elements

2351A, 2361A, 2352A, 2362A, 2500A, 2600A, respectively, of fig. 2A. Further, although the configuration of the power generation unit 2100B in fig. 2B is different from that of the power generation unit 2100A in fig. 2A, some elements of the power generation unit 2100B in fig. 2B are the same as those included in the power generation unit 2100A in fig. 2A. For example, PV strings 2111B and 2112B of fig. 2B are the same as

PV strings

2111A and 2112A, respectively, of fig. 2A. Similarly, DC/AC converter 2130B of fig. 2B is the same as DC/AC converter 2130A of fig. 2A.

The following six (6) steps describe how the power generation unit 2100A is modified into a configuration of 2100B, and the following six (6) steps are described with respect to the left-hand side in fig. 2B. Step 1 is to add a set of decoupling diodes 2311B between the solar strings 2111B and 2112B and the three-phase DC/AC converter 2130B that is practicing the blind MPPT configuration. Step 2 is to add energy reservoir 2410B to the configuration. Step 3 is to connect the energy reservoir 2410B to the DC terminal of the DC/AC converter 2130B through another set of decoupling diodes 2312B and through a switch SW 1. Step 4 is to add another three-phase DC/AC converter 2130S (20kW) to the configuration, the converter 2130S operating according to the designed orientation of the MEUPT controller 2320B. Step 5 is to connect the DC/AC converter 2130S to the energy reservoir 2410B through another set of decoupling diodes 2313B and through switch SW 2. Step 6 is to connect the output terminals of converter 2130S to power and energy measurement instrument sets 2351B and 2361B via switch SW 3. Note that the "set of decoupling diodes" referred to may be those diodes referred to in the art as "blocking diodes". It should also be noted that the addition of switches SW1, SW2, and SW3 as depicted in FIG. 1B allows the relevant devices to be introduced into (or removed from) the experiment at the appropriate time in the design experiment execution step described below.

First night after the above modifications are made; SW1 is on and switches SW2 and SW3 are off.

Transducers

2130B and 2230B begin operating the next morning.

Power meters

2351B and 2352B measuring the two outputs of the

power generation units

2100B and 2200B show the same reading. As indicated by the measurement of the high terminal voltage of reservoir 2410B, reservoir 2410B also begins charging. The system operates as described throughout the day of the first day. The measured energy provided from the two

power generation units

2100B and 2200B is equal; as indicated by readings from kilowatt-

hour meters

2361B and 2362B. This experimental procedure shows that the addition of the decoupling diode set 2311B and the energy reservoir 2410B does not change the power and energy generation of the power generation unit 2100B.

The switches SW1, SW2, and SW3 are turned on during the night (second night) after the first day operation.

Converters

2130B and 2230B begin operating in the early morning (second day), while converter 2130S begins operating at a lower power conversion level approximately 15 minutes after

converters

2130B and 2230B begin operating. Thereafter, converter 2130S increases its conversion power level approximately every 2 minutes; this is consistent with controller design and increments of the accumulator level. The reading of power meter 2351B (for cell 2100B) reaches approximately twice the reading of power meter 2352B (for cell 2200B) throughout the day until near sunset. By the end of the next day, the energy provided from the two

power generation units

2100B and 2200B is derived from readings from two kilowatt-hour meters. The results show that the energy provided from the modified power generation unit 2100B is more than twice the energy provided from the unmodified power generation unit 2200B. During the next six consecutive days, switches SW1, SW2, and SW3 remain on, and the energy provided from modified power-generating unit 2100B per day is always more than twice the energy of power-generating unit 2200B.

The next night, switches SW2 and SW3 are open. In the subsequent 5 consecutive days, the measurement energy supplied from the

power generation units

2100B and 2200B is returned to the same level with the switches SW2 and SW3 kept open. The next night, switches SW2 and SW3 are turned on again. In the next consecutive 5 days, with the switches SW2 and SW3 kept on, the measured energy generation of the power generation unit 2100B again becomes more than twice the measured energy generation of the power generation unit 2200B each day.

As mentioned above, the stepwise execution of this experiment clearly demonstrates the presence of the cited residual energy in the PV power plant as predicted by the cited patent publications (US2016/0036232 and US2017/0149250a 1). Especially in PV power plants, residual energy is still present when the generated DC energy is extracted by a three-phase DC/AC converter. The MEUPT optimizer may capture and utilize this remaining energy to increase the power provided to the grid.

The ninth part: configuration of designed MEUPT optimizer

The modified power generation unit 2100B (as described above and depicted in fig. 2B) can be used as an example of a PV power generation unit that includes a MEUPT optimizer. In this case, the MEUPT optimizer includes: three

decoupling diode groups

2311B, 2312B and 2313B; an energy storage 2140B; and a MEUPT controller 2320B. Note that the decoupling diode group is hereinafter referred to as "decoupling means".

The connection of the MEUPT optimizer module is depicted in FIG. 2B and described above. Note that in this embodiment, the remaining energy is passively extracted by the energy reservoir 2410B. Another power extractor that extracts the remaining energy stored in the energy reservoir 2410B is included as a module in the three-phase DC/AC converter 2130S. The AC power conversion level of converter 2130S is adjusted by MEUPT controller 2320B so that the power charged to energy reservoir 2410B is substantially balanced with the power discharged from energy reservoir 2410B. Thus, the "net" power charged into the energy reservoir over a period of time may be as small as desired. A smaller net power charge has the benefit of allowing a smaller energy reservoir 2410B, but at the expense of tighter control by MEUPT controller 2320B.

Another embodiment is depicted in fig. 3. This embodiment shows a configuration of a PV power plant 3000 incorporating a MEUPT optimizer, the MEUPT optimizer comprising only one AC power generation unit 3100, the AC power generation unit 3100 converting solar energy to DC power using a 500kW solar panel 3110. In other words, the AC power generation unit 3100 includes a DC generator 3110 and a three-phase DC/AC (500kW) converter 3130. The generator 3110 generates DC power using 80 parallel-connected solar strings. Each of the 80 solar strings includes 25 solar panels connected in series; each panel is declared to have 250W DC power generation capability at both noon and clear sky. Note that the DC generator 3110 is referred to as a 500kW electrical power generator (80 × 25 × 250W — 500 kW); and the PV power plant is referred to as a 500kW PV power plant.

As depicted in fig. 3, the generator 3110 supplies DC power to a three-phase DC/AC converter 3130 (declared as 500kW) through a decoupling arrangement 3311. The generator 3110 also supplies DC power to the energy reservoir 3410 through the decoupling device 3312 and serves as a source of DC energy for charging the energy reservoir 3410. Thus, the remaining energy is passively extracted by the energy reservoir 3410. The energy reservoir 3410 then provides (or discharges) DC power to another three-phase DC/AC converter 3130S (declared as 500kW) through the decoupling arrangement 3313. Converter 3130 operates as an MPPT optimizer, while converter 3130S operates as a MEUPT controller. Converters 3130 and 3130S convert the separately supplied DC power to three-phase AC power and deliver it to grid 3600 via the same transformer 3500.

Note that the DC/AC converter used in the above description can be classified into two types; that is, one type receives its DC power directly from the PV solar string, while the other type receives its DC power from the energy storage. When it is necessary to distinguish the type of converter in this disclosure and the following detailed description, one that receives DC power from the PV solar string is also referred to as a "PS DC/AC converter"; while the other receiving DC power from the energy reservoir is also referred to herein as an "ER DC/AC converter". When differentiation is needed in the case of using a three-phase DC/AC converter in the present disclosure, the converters will also be classified and referred to herein as "PS three-phase DC/AC converter" and "ER three-phase DC/AC converter", respectively.

To reiterate on a broader scale; the configuration as depicted in fig. 4 shows that the MEUPT optimizer provides optimized service to an x MWPV power plant with properly arranged strings of solar panels with power generation capacity rated at x MW. The generated DC power is extracted by a "PS three-phase DC/AC converter" 4130 declared by the manufacturer as y MW through a decoupling device 4311. The remaining power is charged into the energy reservoir 4410 through another decoupling device 4312; thereby extracting and storing the remaining energy. The remaining energy stored is then converted by another "ER three-phase DC/AC converter" 4130S, declared as z MW by another manufacturer, through another decoupling device. One of the converters 4130 is regulated by the MPPT optimizer and the other converter 4130S is regulated by the MEUPT controller. Both converters convert an appropriate amount of DC power into three-phase AC power; and provides three-phase AC power to grid 4600 via the same transformer 4500. Note that in this configuration, x ═ y ═ z ═ 0.5.

Fig. 5 depicts another embodiment including a MEUPT optimizer in a large PV power plant. The power plant is equipped with a solar panel string 5110 rated at 0.5MW and two three-phase DC/AC converters 5130 and 5130S declared as 500 kW. This embodiment shows another configuration for the MEUPT optimizer. PV power plant 5000 may be considered to comprise one AC power generating unit (hereinafter also referred to as "AC power generating unit 5100"). The AC power generation unit 5100 includes: a DC generator 5110 comprising a solar panel rated at 500 kW; and two three-phase DC/AC (500kW each stated) converters 5130 and 5130S. The generator 5110 uses 80 parallel-connected solar strings that produce DC power. Each of the 80 solar strings includes 25 solar panels connected in series; each solar panel has a power generation capacity of 250W rated. The energy reservoir 5410 receives DC power from the generator 5110 through the decoupling device 5311. The two three-phase DC/AC converters 5130 and 5130S receive DC power from the energy reservoir 5410 through two separate decoupling devices, including a decoupling device 5312 for the converter 5130 and a decoupling device 5313 for the converter 5130S. The converters 5130 and 5130S are regulated by the MEUPT controller to draw an appropriate amount of power from the energy reservoir 5410 and convert the DC power to three-phase AC power for provision to the grid 5600 via the transformer 5500.

To more broadly illustrate the configuration depicted in FIG. 5: the MEUPT optimizer provides optimization services to x MW PV power stations. The PV power plant has one AC power generating unit with a string of solar panels with a total rated DC power generation capacity of x MW. The DC generator charges the energy reservoir through the decoupling device. The energy reservoir supplies DC power to two three-phase DC/AC converters through two separate sets of decoupling devices. The total conversion capacity of two "ER three-phase DC/AC converters" declared by the manufacturer is z1+ z2 ═ z MW. Both converters are regulated by the MEUPT controller to convert the correct amount of DC power to three-phase AC power. The electricity generated by the two converters is provided to the grid via the same transformer. The configuration described above is being redrawn and is depicted in fig. 6. Note that in this configuration, x is 0.5, y is 0, and z is 1.

The description will now compare the two configurations depicted in fig. 4 and 6. In the configuration depicted in fig. 4, the DC generator supplies DC power to a "PS three-phase DC/AC converter" having a y MW capability as declared by the manufacturer; and charging the remaining power to the energy storage. In fig. 4, the energy reservoir supplies DC power to an "ER three-phase DC/AC converter" having a z MW capability as stated by the manufacturer. Without the "PS three-phase DC/AC converter" (i.e., y 0) in the configuration depicted in fig. 6, all generated DC power is charged to the energy reservoir through the decoupling device; and the energy reservoir supplies DC power to two "ER three-phase DC/AC converters" through two separate sets of decoupling devices. Thus, in the configuration of fig. 4, x-y-z-0.5; in the configuration of fig. 6, x is 0.5, y is 0, and z is 1. In yet another embodiment of fig. 6, the energy reservoir 6410 is absent. In contrast, the solar string 6110 provides DC power to the converter 6130 via the decoupling device 6311.

Now, for the MEUPT optimizer, the only remaining design issue is to identify the optimal power matching relationship between the parameters representing the rated capability of the solar string and the parameters representing the rated capability of the converter. In particular, the task is to identify the best case relationships between x, y and z values. As a reminder, in a conventional PV power plant as described in the second section, the value of the sum y + z is not greater than the value of x.

Note also that the value x is specified as the MW value of the rated DC power generation capability of the PV string; assigning the value y as a total MW value of the capabilities declared by the manufacturer of the "PS three-phase DC/AC converter" that converts the DC energy supplied by the PV string; while the value z is assigned as the total MW value of the capabilities declared by the manufacturer of the "ER three-phase DC/AC converter" which converts the DC energy supplied by the energy reservoir.

For example, in FIG. 6, x is equal to 0.5, the total PV capacity stated by the 0.5MW manufacturer; y is equal to 0, which means that the "PS three-phase DC/AC converter" is not installed; z equals 1, which means that two "ER three-phase DC/AC converters" of total capacity 1MW declared by the manufacturer are included for receiving DC power from the energy reservoir and converting the DC energy into three-phase AC power. Note that in both configurations described above, the value of y + z is not less than 2 times the value of x. The term "capability" is also referred to as the "power rating" of the device; and are interchangeable hereinafter unless otherwise indicated.

The tenth part: the optimum power matching relationship.

The definition of the rated power for a solar panel is different from that of a DC/AC converter due to different disciplines (industries). The rated power of the solar panel is defined as the maximum DC power that the solar panel can produce at midday with clear sky. The solar panel manufacturing industry uses a predetermined type of light (referred to herein as a "standard light") to simulate clear sky; and at noon is simulated by the luminous flux passing through the surface of the solar panel being illuminated perpendicularly. Thus, the power generation capability declared by the manufacturer may be very close to the capability of an actual DC generator. Experiments conducted by the inventors have also confirmed the above statements. Thus, it can be judged that the total DC generating capacity of the PV solar string is reliable; and the title "manufacturer declared capability" is omitted herein in describing the power rating of the solar string. On the other hand, the DC/AC converter manufacturing industry defines the power rating of DC/AC converters according to grid industry conventions (referred to herein as "grid conventions"). This convention and definition of DC/AC converter capability is detailed below.

The AC power grid industry enforces conventions (referred to as grid conventions) to ensure that a constructed three-phase AC power grid can meet declared transmission capabilities. A three-phase AC power grid includes 3 or 4 power lines that can deliver a time-varying sinusoidal function of voltage and current in each pair of power lines as one phase. Grid practice defines the voltage stated in the specification as the "standard" maximum voltage (referred to as the "line voltage") to which the power line is subjected. Also, the specified maximum current stated in the specification is the maximum current to be carried by the power line (referred to as the "maximum phase current"). When manufacturing a device that conforms to grid conventions, the voltage stated in the device specification is the maximum voltage that all relevant components should withstand. Likewise, the maximum current stated in the specification is the maximum current carrying capacity of all relevant components connected to one phase of a pair of power lines. The time-varying functions of the voltage and current of the device also need to conform to the sinusoidal function of each phase in the AC power grid.

Restated, the specified voltage of the three-phase DC/AC converter is defined as the line voltage of the three-phase power; defining a specified maximum current as the maximum current carrying capacity of each pair of power lines in each phase; and the specified maximum power is defined as the sum of the maximum power capabilities that the three phases can withstand. In other words, when complying with grid practice, the power lines and connected power devices of each phase are able to transmit one third of the specified maximum power (1/3), in other words, the "manufacturer's stated rated power" of the three-phase DC/AC converter is 3U I, where U is the phase voltage and I is the phase current. Each pair of power lines is capable of delivering U × I power, 1/3 of "manufacturer declared rated power"; and each module connected to the pair of power lines must also carry or deliver 1/3 for the specified power rating stated, when in compliance with grid practice.

For example, the designation "AC voltage 315V AC; maximum current is 916 amperes; and the maximum power output is 500kW "for a three-phase DC/AC converter as an example. The specification "AC voltage 315V AC" should be understood as: "the output line voltage of this converter is 315 volts". Alternatively, in three-phase balancing, the phase voltage U for each phase is 315/1.732 181.9 volts (where 1.732 is the square root of 3, i.e., the ratio of line voltage to phase voltage). The designation "maximum current 916 amps" should be understood as: the power lines and all components in each phase are designed to ensure a current carrying capacity I of 916 amps. The specified "maximum power output of 500 kW" is to be understood as: maximum power conversion and delivery capacity U I181.9 916 500/3KW for all components of each DC/AC conversion phase; and the maximum total power conversion and delivery capacity of the relevant module in the three converted phases is the sum of each phase, i.e. 3U I181.9 916 500kW, which is defined as "manufacturer declared rated power" 3U I when complying with the grid convention set out in the preceding paragraph.

The three phases in a three-phase DC/AC converter are strictly correlated to have a phase difference of 120 °. In other words, a pair of power lines (phases) delivers UxI sin²(ω t) time-varying power; while the second phase delivers Ux I sin²(ω t +120 °) time-varying power; and the third phase delivers Ux I sin²(ω t-120 °) time-varying power. Each pair of three-phase power lines delivers three oscillating AC power trains related to each other with strict correlation. Note that the power conversion capacity p (t) is not equal to the defined "manufacturer's stated power rating". The power conversion capacity p (t) is expressed as a function of time and is derived from the defined three-phase AC power limit.

In other words, the DC/AC power conversion capacity p (t) is derived from the sum of the time-varying power outputs of the three phases; where the phase difference is 120 ° strictly correlated; and wherein the power waveform is in sin²(ω t) or cos²(ω t) square sine oscillations; and is synchronized (same phase and frequency) with the grid, forcing the angular frequency ω to remain constant.

Now, the time-varying power conversion capacity p (t) of the three-phase DC/AC converter is derived. The power conversion capacity of a three-phase DC/AC converter as a function of time is p (t) ═ U × I (sin)²(ωt)+sin²(ωt+120°)+sin²(ω t-120 °)). As defined above, U is the phase voltage, I is the phase current, and ω is the constant angular frequency of the grid. Also, sin may be shown²(ωt+120°)+sin²(ωt-120°)＝cos²(ω t) + 1/2. Thus, the power conversion capacity p (t) of a three-phase DC/AC converter as a function of time is p (t) U I (sin)²(ωt)+sin²(ωt+120°)+sin²(ωt-120°))＝U*I*(sin²(ωt)+cos²(ωt)+1/2)＝U*I*(1+1/2)＝3/2(U*I)。

In other words, the sum of these strictly correlated three pulsed power trains in the three phases is constant. In other words, the total power output of the three pairs of power lines is constant. Or the sum of the three modules associated with the three phases is constant. However, the constant is only equal to half of the "declared power capability" (1/2). This is the relationship between the power conversion capacity of a three-phase DC/AC converter and the defined "declared power capability" when complying with grid conventions.

Note that as previously mentioned, the "manufacturer declared power rating" or referred to as the "manufacturer declared power capability" of a three-phase DC/AC converter is 3U I when grid conventions are met. Comparing this with the power conversion capacity p (t) ═ 3/2(U × I) derived above; it is evident that the derived DC/AC power conversion capacity of a three-phase DC/AC converter is only half of the "manufacturer's stated power capability".

As an example, the designation "AC voltage 315V AC" described above is again employed; maximum phase current 916A; and the maximum power output is 500kW "for a three-phase DC/AC converter as an example. In practice, the DC/AC power conversion capacity of the three-phase DC/AC converter is only 250 kW. To derive the above conclusions, it was first confirmed that the maximum power asserted, 500kW, was indeed equal to 3U I, where U is the phase voltage derived from the specified line voltage and I is the maximum current asserted; the power conversion capacity of the converter is equal to 3/2U I250 kW.

The best power matching relationship for the parameters x, y and z (as defined) is that the value of (y + z) should not be less than 2 x. Where the relevant PV power plant comprises an x MW PV solar string; wherein the total "manufacturer declared power capability" of the "PS three-phase DC/AC converter" is y MW; and wherein the total "manufacturer declared power capability" of the "ER three-phase DC/AC converter" is z MW. The "PS three-phase DC/AC converter" and the "ER three-phase DC/AC converter" may be operated by one or more MPPT controllers or by one or more MEUPT controllers. To practice MEUPT optimization, all DC/AC converters are preferably operated by the MEUPT controller.

The eleventh part: summary of the invention

Fig. 7 shows the configuration of a PV solar power plant 7000 in an abstract manner. The power plant includes a total of x MW solar panels arranged in a solar string 7100. DC power generated in solar string 7100 provides DC power input to a set of three-phase DC/AC converters 7301 through decoupling device 7201; and the surplus power is charged into the energy storage 7400 through the decoupling device 7202. The energy reservoir 7400 provides a DC power input to a set of three-phase DC/AC converters 7302 through a decoupling device 7203. Both three-phase DC/

AC converters

7301 and 7302 provide the converted three-phase AC power to the grid 7600 through a transformer 7500. The total "manufacturer declared capability" of converter 7301 is y MW. The total "manufacturer declared capability" of the converter 7302 is z MW. The value of the sum of (y + z) is not less than the value of 2 x. Note that when a conventional PV power plant is described using a similar configuration as described in the second section, the value of (y + z) is not greater than the value of x. Thus, in designs with a value of (y + z) greater than x or even better 1.1 times x; this means that some surplus energy can be captured to enhance the electrical energy provided to the grid.

Both

converters

7301 and 7302 may be operated by the MEUPT controller described above. In some embodiments, some, one, or even none of the converters are operated by the MEUPT controller. Furthermore, in some embodiments, one or some of

decoupling devices

7201, 7202, and 7203 may be omitted from the configuration. PV solar string 7100 provides DC power input to converter 7301. Accordingly, converter 7301 is referred to herein as a "PS converter". The energy reservoir 7400 provides a DC power input to the converter 7302. Therefore, converter 7302 is referred to herein as an "ER converter". The terms total "manufacturer stated power rating" and total "manufacturer stated power capability" shall be abbreviated herein as "declared power".

To reiterate the description of the configuration depicted in fig. 7: PV power plant 7000 comprises a solar string 7100 of x MW as a DC generator. DC generator 7100 provides input through decoupling 7201 to "PS converter" 7301 with "declared power" of y MW; and the remaining power is charged to the energy reservoir 7400 through another decoupling device 7202. The energy reservoir 7400 provides input to an "ER converter" 7302 having an "asserted power" of z MW through a decoupling device 7203. All three-phase DC/

AC converters

7301 and 7302 provide the converted three-phase AC power to the grid 7600 through a transformer 7500. In some embodiments, the value of (y + z) is not less than the value of 2 x. However, when the value of (y + z) is greater than the value of x, the design may receive partial benefit to enhance the sale of electrical energy to the grid.

A MEUPT optimizer according to the principles described herein may serve a small PV power plant or a large PV power plant comprising one or more AC power generating units. Furthermore, by means of a suitably designed decoupling device, energy leakage from the energy reservoir through the PV solar string can be prevented. Furthermore, the phenomenon of "mutual power annihilation" that is found can be prevented by appropriately designed decoupling means. Likewise, the energy reservoir may be used to receive only the remaining energy after energy extraction by the "PS converter", or all of the generated DC energy before any extraction. Finally, the MEUPT optimizer may also serve PV power plants equipped with single-phase DC/AC converters.

The twelfth part: design constraints for MEUPT controllers

Fig. 8 shows a MEUPT controller 8000 (also referred to as a "system controller") that represents an example of the MEUPT controller 2320B of fig. 2B. The MEUPT controller 8000 includes 3 executable components: a detection component 8100, a determination component 8200, and a delivery component 8300.

The detection component 8100 measures the level of stored energy in the energy reservoir 8400. Examples of the energy storage are the energy storage 2410B of fig. 2B, the energy storage 3410 of fig. 3, the energy storage 4410 of fig. 4, the energy storage 5410 of fig. 5, the energy storage 6410 of fig. 6, and the energy storage 7400 of fig. 7.

The determination component 8200 determines an appropriate power draw level such that the charge supplied to and discharged from the energy reservoir 8400 approaches equilibrium.

Delivery component 8300 delivers the encoded message of the appropriate power draw level determined above to remaining DC/AC converter 8500. The converter interprets the encoded message and complies with the encoded message such that the converter can operate continuously at a directed power level to bring the charging energy close to equilibrium. Examples of transducers 8500 drawn from the energy reservoir 8400 are transducer 2130S of fig. 2B, transducer 3130S of fig. 3, transducer 4130S of fig. 4, transducer 5130S of fig. 5, transducer 6130S of fig. 6, and transducer 7302 of fig. 7.

To derive the MEUPT economic optimizer, the design of the MEUPT controller takes into account the following parameters and variables: (1) capacity of energy reservoir 8400; (2) the rise/fall speed of the DC/AC converter 8500; (3) I-V characteristics of the solar string; (4) climate of the location of the PV power plant; and (5) the ability of the MEUPT controller to work with the remnant DC/AC converter minimizes (or balances) the difference between the amount of charge provided to the energy reservoir and the amount of charge drawn from the energy reservoir. A simple design can only be derived if a custom designed controller is applied to each and every PV power plant taking into account these parameters and variables.

The thirteenth part: MEUPT controller design

Custom designing a MEUPT controller for each and every PV power plant to use it is impractical. On the other hand, it is very difficult to pursue a simple design for the required MEUPT controller; particularly where custom design of the controller is not allowed. However, the terminal voltage of the energy reservoir can be considered as a measure that is influenced by each of the 5 parameters and variables. Thus, the above 5 design constraints can be broken down into two parts when selecting the terminal voltage of the MEUPT energy reservoir as the determining parameter.

When comparing the measured terminal voltage to a set of site-specific "standard voltage intervals"; the inventors have appreciated that when the power extraction level is (1) too low, (2) too high, or (3) just, the power extraction and conversion levels currently performed by the system may be quantified. Thus, the design task of a MEUPT controller can be decoupled from 1) a generic industrial controller, plus 2) a custom-constructed site-specific "standard voltage interval" table (referred to herein as "Voltage interval meter”)。

Once the site specific voltage interval table is constructed for the PV power plant; the voltage interval table may work in conjunction with the industrial controller to accomplish the desired functionality of the MEUPT controller. The industrial controller then comprises a detection component, a determination component and a delivery component as also shown in fig. 8. In this case, however, the detection part 8100 measures the terminal voltage of the energy reservoir 8400. The determination section 8200 compares the measured voltage with the voltage interval table; and determines an appropriate amount of power draw to bring the charging energy close to equilibrium. Delivery component 8300 again delivers the encoded message of the appropriate power draw level determined above to the remaining DC/AC converter; such that the converter can be operated continuously at a directed power level to bring the input and output charges of the accumulator 8400 near equilibrium.

In one embodiment, the detection component 8100 of the MEUPT controller 8000 measures the terminal voltage of the remaining energy reservoir 8400 in real time. Even so, the determination component 8200 can perform a comparison (of the measured voltage versus the voltage interface table) in each specified time interval comparison. This comparison may result in one of three situations:

(1) if the comparison of the measured voltage to the voltage interval table indicates that the power level is too low, then the controller 8000 may request (via delivery component 8300) that the three-phase DC/AC converter 8500 increase one level of power extraction and conversion for the next specified time interval;

(2) if the comparison of the measured voltage to the voltage interval table indicates that the power level is too high, the controller 8000 may request (via delivery component 8300) that the three-phase DC/AC converter 8500 reduce one stage power draw and conversion for the next specified time interval;

(3) if the comparison of the measured voltage to the voltage interval table indicates that the power level is correct, the controller 8000 may request that the three-phase DC/AC converter 8500 maintain the same power draw level for the next specified time interval, at least until the next comparison occurs.

When the power extraction/conversion regulation level of the DC/AC converter is sufficiently small, the above design can be adapted to a wide variety of energy storage capacities; up/down ramp speeds for a variety of DC/AC converters; the method is suitable for the I-V characteristics of various solar strings; and all climates applicable to PV sites. It is therefore important that the controller can direct the three-phase DC/AC converter to draw a small adjustment step of power from the energy reservoir.

Typical conventional centralized three-phase DC/AC conversionThe device can operate with very small adjustment steps when it is booted. However, the equipped communication channel, known in the art as "Dry connection box"(and referred to as such herein), typically has only a 6-bit communication channel via optical messages. To command more than 6 power draw levels through the dry connect box, codec techniques are employed. This technique allows delivery of up to 2⁶64 messages to command the power extraction level. By adjusting the power extraction level by as much as 64, the desired net balance of near zero in input energy and output energy of the energy reservoir can be technically achieved.

The fourteenth section: PV power plant comprising a MEUPT optimizer

As shown in fig. 9, a PV power plant 9000 comprises a MEUPT optimizer 9200, and the MEUPT optimizer 9200 comprises a MEUPT controller 9210. The MEUPT controller 9210 includes 3 executable components: that is, the detecting means 9211 for measuring the terminal voltage of the surplus energy reservoir 9400; a determining component 9212 for comparing the measured voltage with a voltage interval table of the PV station; and a delivery part 9213 for notifying the three-phase DC/AC converter 9502 of start-up, drop, or maintenance via the delivery part 9213.

Parts

9211, 9212 and 9213 of fig. 9 are examples of

parts

8100, 8200 and 8300 of fig. 8, respectively. The energy reservoir 9400 of fig. 9 is an example of the energy reservoir 8400 of fig. 8. Converter 9502 is an example of converter 8500 of fig. 8.

PV power plant 9000 also comprises a PV solar string 9100. The solar string 9100 converts solar energy into electricity; and delivers the generated DC power to the remaining energy reservoir 9400 through the decoupling device 9320. The three-phase DC/AC converter 9502 receives DC power input from the residual energy reservoir 9400 through a decoupling device 9330. Solar string 9100 of fig. 9 is generally a DC energy source for charging an energy reservoir and is an example of solar string 2111A and 2111B of fig. 2B, solar string 3110 of fig. 3, solar string 4110 of fig. 4, solar string 5110 of fig. 5, solar string 6110 of fig. 6, and solar string 7110 of fig. 7. The decoupling 9320 of fig. 9 is an example of the decoupling 2312B of fig. 2B, the decoupling 3312 of fig. 3, the decoupling 4312 of fig. 4, the decoupling 5311 of fig. 5, the decoupling 6311 of fig. 6, and the decoupling 7202 of fig. 7. The decoupling device 9330 of fig. 9 is an example of the decoupling device 2313B of fig. 2B, the decoupling device 3313 of fig. 3, the decoupling device 4313 of fig. 4, the decoupling device 5313 of fig. 5, the decoupling device 6313 of fig. 6, and the decoupling device 7203 of fig. 7.

As described above, MEUPT controller 9210 directs three-phase DC/AC converter 9502 to draw the appropriate amount of energy from energy reservoir 9400 to balance the input energy charged from solar string 9100; this causes the energy charged or drawn into the energy reservoir 9400 to approach zero. Thus, a smaller energy reservoir 9400 is sufficient for the PV plant. The converted AC power from the DC/AC converter is provided to the connection grid 9700 through a transformer 9600.

As used herein, the term "executable component" is used with respect to fig. 8 and 9. The term "executable component" is the name given to structures well known to those of ordinary skill in the computing arts, which structures may be software, hardware, firmware, or a combination thereof. For example, when implemented in software, those of ordinary skill in the art will appreciate that the structure of an executable component may include software objects, routines, methods that may be executed on a computing system, whether or not such executable component exists in a heap of the computing system, or whether or not the executable component exists on a computer readable storage medium.

In such cases, those of ordinary skill in the art will recognize that the structure of the executable components reside on computer-readable media such that, when interpreted by one or more processors of the computing system (e.g., by a processor thread), cause the computing system to perform functions. Such structures may be computer-readable directly by the processor (as in the case where the executable components are binary). Alternatively, the structure may be structured to be interpretable and/or interpretable (whether in a single stage or in multiple stages) to generate such a binary that is directly interpretable by the processor. When the term "executable component" is used, such an understanding of an example structure of an executable component is well within the understanding of one of ordinary skill in the computing arts.

The term "executable component" is also well understood by those of ordinary skill in the art to include structures that are implemented exclusively or near exclusively in firmware or hardware, such as within a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), or any other special purpose circuit. Thus, the term "executable component" is a term for structures well known to those of ordinary skill in the computing arts, whether implemented in software, hardware, or a combination.

The present invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.

Claims

1. An energy storage system comprising:

an energy reservoir charged by DC energy from a DC energy source while releasing the DC energy to a DC/AC converter;

a system controller for regulating the DC energy released from the energy reservoir to the DC/AC converter to bring the amount of DC energy charged to the energy reservoir near equilibrium, the system controller comprising:

a detection component configured to measure a level of stored energy in the energy reservoir;

a determination component configured to use the measured energy storage level of the energy level to evaluate whether an adjustment is to be made; and

a delivery component configured to encode an instruction to perform the adjustment into an encoded message when the determination component determines that the adjustment is to be made, and further configured to deliver the encoded message to the DC/AC converter.

2. The energy storage system of claim 1, the detection component measures the stored energy level by measuring a terminal voltage of the energy reservoir.

3. The energy storage system of claim 2, the measurement of the terminal voltage of the energy reservoir occurring in real time.

4. The energy storage system of claim 3, the determination component configured to perform an evaluation of whether to adjust periodically at specified time intervals.

5. The energy storage system of claim 1, the determination component configured to perform an evaluation of whether to adjust periodically at specified time intervals.

6. The energy storage system of claim 1, the delivery component configured to deliver the encoded message to the DC/AC converter through a dry-connect box.

7. The energy storage system of claim 1, the delivery component configured to deliver a respective encoded message to each of a plurality of DC/AC converters including the DC/AC converter.

8. The energy storage system of claim 1, the delivery component configured to deliver the respective encoded message to at least one of the plurality of DC/AC converters via the dry junction box.